This study will use a set of validated pediatric neurodevelopmental examinations to assess the longitudinal stability and predictive power of neurodevelopmental findings in normal and learning-disabled students of at least average intelligence. We expect to find that neurodevelopmental task attainment has predictable rates and patterns. We intend to use our longitudinal data about the developmental attainment of children to construct a set of "developmental growth curves" which will allow us to predict for parents and teachers how a child will progress over time. Finally, we hope to use our data to examine the relationship between neurologically-based developmental attainment and learning disabilities subtypes. Subjects will be recruited from four study sources: learning- disabled patients will come from the School Performance Consultation Program, a multidisciplinary clinic for children with school problems, while normal patients will be recruited from a hospital-based group practice, a community-based group practice, and if needed, the Boston Public Schools. Informed consent will be obtained. Data on medical events, educational, neurodevelopmental and neuropsychological performance, IQ, SES and family history will be obtained from 100 learning-disabled and 50 normal subjects. These individuals will be followed 18 months and 3 years after their initial clinical evaluation, and neurodevelopmental, neuropsychological and educational measures will be repeated. Data will be analyzed using a longitudinal two-way design, cluster analysis, analysis of variance and other correlational measures. Statistical power calculations indicate that our sample size will be large enough to detect correlations in which the size of the effect is 0.25 or greater. We believe that, with measurements every 18 months, we will be seeing effects of at least this size. The results of this study should provide us with valuable information about the patterns of developmental growth in normal and learning-disabled children. In addition, we hope to demonstrate the importance of the neurologic-developmental contribution to a child's academic success. If we are able to predict future function of children based on these longitudinal data, this will enable us to help parents, teachers and others who work with learning-disabled children to develop more effective strategies in planning educational interventions.